Complicated intra-ABdominal Infections (CABIs) are abdominal infections where there is an abscess inside the abdomen, or a hole (perforation) in an abdominal organ such that infected material e.g. faeces, leaks into the abdominal cavity. A recent review of CABIs after gut surgery found that they can occur in several ways. They can occur in different parts of the abdomen, can be different sizes, and may or may not be caused by a perforated bowel. Management includes, where possible, surgical drainage of an abscess or treatment of the damaged bowel. In addition, all patients are given antibiotic therapy. Despite the varied ways that CABIs occur, we currently tend to treat all CABIs with antibiotics in a similar way. CABIs are associated with significant morbidity and mortality. Despite a significant amount of disease there is little clinical evidence with which to base treatment on. One research study evaluated a short course of antibiotics (4 days) compared with a longer course (up to 10 days) in combination with surgical removal of infection. There was little difference in outcomes, but in both groups about 1 out of every 7 patients had a relapse. A recent review of patients with CABI in Leeds, not in a research study and where surgical removal infection is uncommon and antibiotic durations were short, showed that the risk of relapse was even higher (about 1 in every 3 patients). The antibiotic management of CABIs in the UK is variable and involves giving between 4 days to 28 days of antibiotics.

In summary, there is an unacceptably high relapse rate in patients treated for CABI, and uncertainty about the best length of antibiotic therapy that should be used to prevent these relapses. We therefore propose to investigate if long course antibiotic therapy (28 days) is more effective than short course antibiotics (≤10 days) in preventing relapses of CABI.

Complicated intra-abdominal infection (CABI) extends beyond the hollow viscus of origin into the peritoneal space and is associated with either abscess formation or peritonitis [Solomkin 2010]. CABIs are heterogeneous in aetiology and include spontaneous infections arising from a perforated viscus, including the stomach, biliary tree, colon, appendix and reproductive organs. In addition there are post-operative complications such as peritonitis, abscess, and failure of surgical anastomoses. CABIs are also heterogeneous in their characteristics. A recent review of organ space Surgical Site Infections (SSI-Os) after colorectal surgery demonstrated that infections were varied with regard their location, as well as size and number and nature of collections, and presence of an ongoing source e.g. anastomotic breakdown (Rothwell 2016). Despite the varied origin of these infections, there are similar management strategies which centre on the source control, e.g. drainage of intra-abdominal fluid collections, and administration of antibiotic therapy. CABIs are a common in-hospital clinical challenge, in part due to the varied pathology they are caused by, and are associated with significant morbidity and mortality (DeFrances 2005, Brun-Buisson 1995). Despite this burden of disease, there is little clinical evidence with which to base treatment on, even before taking account of their heterogeneous nature. One study of CABI following source control procedures evaluated a short course of antibiotics (4 days) compared to a course of up to 2 weeks after clinical and biochemical improvement (up to 10 days). Whilst there was little difference in outcomes, both groups had a high relapse rate of approximately 15%. In addition, it is common for source control procedures not to be completed in routine clinical practice, being completed in only 17% of Leeds patients with post-operative CABI (Rothwell 2016). In this Leeds data of post-operative CABIs, where source control rates are low relapse rates were 40% (Rothwell 2016). For CABI infections, standard UK management is variable and involves between 4 and 42 days' antibiotics. This indicates an unacceptably high relapse rate in patients treated for CABI where source control is and is not undertaken. We therefore propose to investigate if long course antibiotic therapy (28 days) is clinically more effective than short course antibiotics (4 days) in preventing relapses of CABI. This study will include patients who have and have not undergone source control procedures which will be completed as per standard practice.

Complicated intra-abdominal infections require source control when possible, e.g. a surgical procedure to remove an infection, and antibiotic therapy to obtain cure. Source control is not always possible. CABIs are associated with mortality and prolongation of hospitalisation. After apparently effective treatment, potentially including source control procedures and antibiotics, infections can relapse. There are a number of reasons for relapse; one is that antibiotic treatment may not have been given for long enough to eradicate the bacteria from, what should be, a sterile intra-abdominal cavity. Antibiotics are given until a patient is better, but not until all bacteria are eradicated, allowing them to regrow and re-start an infection. Standard antibiotic duration is variable: some doctors provide long courses and others short. We therefore want to compare durations, to see if longer courses of antibiotics are able to help prevent these relapses, or if shorter courses are as effective but have fewer side effects. We have not identified closely related strategies which may optimise the management of CABIs.

The research hypothesis is therefore: In patients with CABI, regardless of source control intervention, there will be a lower relapse rate when treated with 28 days of antibiotics compared to ≤10 days of antibiotics.

Inclusion Criteria:

• Adults >18 years
• The diagnosis of a definite CABI
• Capable of giving informed consent
• No practical or clinical barriers to consuming 28 days of antibiotic therapy, which may include consumption of antibiotics at home

Exclusion Criteria:

• a CABI diagnosed within the previous year
• a CABI diagnosed >6 days prior to screening
• uncomplicated cholecystitis/cholangitis/gall bladder empyema (no perforation or extra-biliary abscess
• a skin and soft tissue infection/abscess not communicating with the peritoneal space
• primary complicated or uncomplicated appendicitis managed surgically
• intra-abdominal infection associated with pancreatitis, pelvic inflammatory disease, primary (spontaneous) bacterial peritonitis (SBP), continuous ambulatory peritoneal dialysis peritonitis (CAPD peritonitis) and Clostridium difficile infection. concurrent infection requiring more than 10 days of therapy
• Infection with a highly resistant bacterium such that antibiotic treatment is considered to be a significantly sub-optimal by the treating microbiologist e.g. multi-resistant carbapenemase producing Entrobacteriacea